Developing Dual Energy CT Applications for Pediatric Imaging Zhu, Xiaowei, MS, DABR, DABMP

Transcription

1 Developing Dual Energy CT Applications for Pediatric Imaging Zhu, Xiaowei, MS, DABR, DABMP Department of Radiology The Children's Hospital of Philadelphia Perelman School of Medicine The University of Pennsylvania

2 Disclosures: Nothing to disclose

3 Background Dual Energy CT (DECT) technology is available on scanners from several vendors and offers significant advantages over classic single energy CT technology in multiple clinical applications. This technology utilizes photon spectra generated by two different tube voltages to identify and distinguish different materials in the body such as iodine, uric acid and calcium.

4 Background Many studies have detailed DECT applications in adults and several have evaluated the relative radiation dose performance of DECT in adult imaging. Little has been published on DECT imaging in the pediatric population and the relative dose performance of DECT imaging in the pediatric population is not well described.

5 DECT Basics Total X-ray Absorption J Fornaro et al. Insights Imaging (2011)2: Materials can be distinguished more easily when large differences in atomic numbers are present. Differences in attenuation curves are magnified at two different energy levels, allowing differentiation on the basis of two CT numbers.

6 DECT Basics Hardware J Fornaro et al. Insights Imaging (2011)2: Rapid voltage switching X-ray tube voltage modulates rapidly to different kvp. Dual-layer detector Superimposed energy sensitive layers detect different energy levels. Dual source dual energy Two X-ray tube with corresponding detectors operate at different kvp with an angular off-set. Dual-energy data are acquired simultaneously.

7 DECT Applications Current clinical applications can be divided in to two groups: Differentiation of materials Bone removal (differentiation of calcium and iodine) Urinary stone analysis (differentiation of uric-acid containing stones, vs. other stone types) Identification of contrast material Reconstruction of virtual non-contrast images from Iodine-enhanced image by subtraction of quantified iodine

8 Dose Neutrality? Manufacturers suggest o DE scans can be performed "dose neutral*" (*only in adult?) A question should be answered by pediatric facilities based on their current SE scan protocols, especially, comparing with already optimized SE protocols. Phantom studies to investigate dose neutrality o Understand the relative dose differences. o Develop DE scan protocols based on our optimized SE protocols

9 Dose Neutrality? Our study 3 (sizes) anthropomorphic phantoms of 1(10kg), 5(19kg), & 10y(32kg) olds with contrast inserts DE scans with matching reference CTDI vol of SE o Head o Chest o Abdomen Scan Modulations o mas and kv (chest, abdomen 0, head: semi) for SE o mas for DE Ref. CTDI vol = Noise Index Eff. CTDI vol = Avg. CTDI Second DE scans with adjusted Ref CTDI vol o Scaled to ratio of Effective CTDI Vol of DE and SE scans

10 The Phantoms Phantom and insert oinserts with 1-6 mm contrast targets o Simulating blood in brain, nodule in chest, and iodine in abdominal regions ofor visual assessments & quantitative analysis

11 Studies of CTDI and CNR Effective CTDI* of SE and DE were compared for o Relative dose differences o In each anatomical region o In each phantom size group * vender use 32cm phantom to report trunk region for all sizes, and 16cm for head region. SE DE CNR Estimated for o All phantom sizes and regions o Ten 3mm slices with the contrast insert in each region Two radiologists provided visual assessments

12 SE, DE Compared

13 Dose Neutral?

14 How about CNR? On CNR analysis of image quality, in head and abdominal regions, there is no statistically significant changes in CNR of dual energy scans when compare with the images of single energy scans for all three phantoms sizes (p-value: 0.06, 0.06, 0.16, for 1y, 5y, 10y). CNR in images processed with DE specific Q40 soft tissue algorithm were slightly better than standard I41f algorithm in abdominal images, judging by the slightly higher p-values in each phantom group.

15 How about CNR? In chest region, although DE Eff CTDI Vol are higher in 1-yearold and 5-year-old phantoms (120%, 111% that of SE Eff CTDI Vol ), there is no significant difference in CNR in DE and SE images in all phantom groups (p-value: 0.11, 0.49, 0.46 for 1y, 5y, 10y). Further forced reduction of Eff CTDI Vol by reducing Ref CTDI Vol (in 1y and 5y) in chest produced DE images that have statistically significant lower CNR for the 1-year-old phantom (p-value 0.007); but no significant CNR reduction was observed in the 5-year-old phantom (p-value:0.31).

16 Practice Recommendation

17 SAMS #1: Dual Energy CT is based on the concept that... a. Materials can be distinguished more easily when large differences in atomic numbers are present. b. Materials can be distinguished more easily when large differences in neutron numbers are present. c. Differences in attenuation curves are magnified at two different energy levels, allowing differentiation on the basis of two CT numbers. d. a) & b) 1% 0% 12% 7% 79% e. a) & c) a. b. c. d. e.

18 Answer to SAM QUESTION 1 Correct Answer, e). Materials can be distinguished more easily when large differences in atomic numbers are present. Differences in attenuation curves are magnified at two different energy levels, allowing differentiation on the basis of two CT numbers. Ref: J Fornaro et al. Dual- and multienergy CT: approach to functional imaging Insights Imaging (2011)2:

19 SAMS #2: Current applications can be divided in to two groups: 22% a. Bone removal and Stone identification 10% 4% 54% 9% b. Contrast removal and Bone removal c. Metal artifact reduction and bone removal d. Differentiation of materials and identification of contrast material e. Stone identification and metal artifact reduction

20 Answer to SAM QUESTION 2 Correct Answer: d) Current applications can be divided in to two groups: Differentiation of materials Bone removal (differentiation of calcium and iodine) Urinary stone analysis (differentiation of uric-acid containing stones, vs. other stone types) Identification of contrast material Reconstruction of virtual none-contrast images from Iodineenhanced image by subtraction of quantified iodine Ref: J Fornaro et al. Dual- and multi-energy CT: approach to functional imaging Insights Imaging (2011)2:

21 DECT Applications at CHOP GU Imaging: CT Urolithiasis > 1 year old Contrast Renal CT for stone and obstruction Virtual Non-contrast (VNC) CT Cystography and Urography (VNC) Oncological Imaging: > 1 year old Abdominal tumor with suspicion of calcifications (VNC) Vascular Imaging: Catheter remnant in vessel (VNC) Arterial Calcinosis (VNC) MSK Imaging: CT extremity with hardware Neuro Imaging: CT spine with hardware CTA of head bone subtraction Diagnosis or post-intervention assessment for bleeding vs. enhancement (VNC)

22 Clinical Cases Renal Stone Composition Color coded red indicating a lower dual energy ratio seen with uric acid based renal calculi. Color coded with blue indicating a larger dual energy ratio seen with calcium (Adolescent male with Lesch-Nyhan syndrome and uric acid renal calculi.)

23 Clinical Cases - CTA with Bone Removal MIP image after bone removal from rotating data set ICA terminus aneurysm Axial bone removal image

24 Clinical Cases Bone Removal Basilar artery aneurysm MIP image after bone removal from rotating data set generated by CT tech at the scanner Volume rendering generated in 3D lab

25 Clinical Cases Bone Removal Kawasaki s Disease Large M1 segment left ICA aneurysm Iodine map image from rotating data set generated at the scanner

26 Clinical Cases - Virtual Non Contrast hemorrhage Large parenchymal hemorrhage secondary to an AVM. Virtual non-contrast confirms the presence of hemorrhage, excluding an enhancing mass.

27 Clinical Cases - Virtual Non Contrast Contrast DECT scan VNC VNC generated from a single phase, contrast enhanced sequence. Note the difference in HU in the left kidney.

28 Clinical Cases Delayed nephrogram from distal left ureterolith Virtual Non-Contrast Incomplete removal of contrast Incomplete iodine removal contrast in renal pelvis Contrast True non-contrast Virtual non-contrast

29 Clinical Cases Virtual Non Contrast True Non-contrast Virtual Non-contrast Subarachnoid hemorrhage Subarachnoid hemorrhage in the prepontine cystern from a basilar anerysm is subtle and difficult to appreciate on the VNC image.

30 Clinical Cases Metallic Artifact Reduction Posterior cervical fusion DE SE DE Scoliosis with posterior spinal fusion

31 Summary The pediatric phantom study Doses are neutral in head and abdominal studies; and are also neutral in chest studies for patients larger than a typical one year old. A lower mas floor exists, when both tube currents reach nominally 10mAs per rotation, thus could result a limitation on dose neutrality when converting SE protocols to DE protocols for very small patients, or from most aggressively dose-reduced SE protocols.

32 Summary The pediatric phantom study Has helped us better understand the dose consequences of converting our single energy protocols to dual energy protocols. Has provided clear clinical guidelines as we begin to implement clinical dual energy protocols in pediatric imaging. We predict a wide range of the dual energy applications in pediatric imaging that are well justified in dose and offer great potential in diagnostic imaging and image quality improvement.

33 Summary CHOP has developed routine clinical applications in metallic artifact reduction renal stone evaluation bone removal in head/neck CTA, and virtual non-contrast in the evaluation of abdominal tumors with suspicion of calcifications

34 Summary Armed with improved understanding, we have continued to develop more clinical applications to take full advantage of dual energy technology in the CT imaging of our pediatric population. believe DECT offers unique advantages in clinical CT imaging and are well justified in doses

35 SAMS #3: Identify the following various dual energy applications, from left to right: 8% a. Metal artifact removal, Liver VNC, Head/Neck CTA bone removal, Abd CTA 6% 3% 81% 2% b. Bone removal, Stone Evaluation, Liver VNC, Head/Neck CTA c. Abd CTA Metal artifact removal,, Head/Neck CTA bone removal, Liver VNC d. Metal artifact removal, Stone Evaluation, Head/Neck CTA bone removal, VNC in abdomen e. None of above

36 Answer to SAM QUESTION 3 Correct Answer: d) (Ref: McCullough W, Vossough A, Darge K, Zhu X, Servaes, S; Implementation of Dual-Energy CT in Pediatric Imaging, Society for Pediatric Radiology Annual Meeting, April 27 May 1, 2015, Bellevue, Washington.) CHOP has developed routine clinical applications in metallic artifact reduction renal stone evaluation bone removal in head and neck CTA, and virtual non-contrast in the evaluation of abdominal tumors with suspicion of calcifications

37 Developing Dual Energy CT Applications for Pediatric Imaging Zhu, Xiaowei, MS, DABR, DABMP Department of Radiology The Children's Hospital of Philadelphia Perelman School of Medicine The University of Pennsylvania